supporting information - university of...

Supporting InformationGold et al. 10.1073/pnas.1014400108SI MethodsExpression and Purification of AKAP79 Complexes. Human cAMP-dependent protein kinase (PKA) RIIα(1–45), mouse CaM and abicistron containing human Ca2þ∕calmodulin (CaM)-dependentprotein phosphatase (PP2B) A-B(5–170), adapted from (1), wereeach cloned into pGEX6P1 for expression in Escherichia coliBL21 cells by induction with 0.5 mM IPTG after the bacteriahad reached OD600 nm of 0.5. Cells were harvested after overnightgrowth at 18 °C. A-kinase anchoring protein 79 (AKAP79) wasexpressed in Sf21 insect cells using the Bac-to-Bac BaculovirusSystem (Invitrogen). Insect cells were harvested after 72 h inmedium containing 10% fetal bovine serum.

The protocols for purification of both the AKAP79–D∕D andAKDPC (AKAP79, PKAD∕D, PP2B, CaM) complexes are initiallyidentical. In the absence of prior association to RII D∕D, AKAP79accumulated into soluble aggregates, as evidenced by gel filtration.Therefore, in both cases, affinity of AKAP79 to GST–D∕D wasperformed as the first purification step. The pellet from 1.5 L ex-pression of D∕Dwas lysed in 100mL lysis buffer A [25 mMTris pH7.5, 0.5 M NaCl, 2 mM DTT, 0.5 mM EDTA, one EDTA-free pro-tease inhibitor tablet (Roche), 5 mgDNase 1, 10 mg lysozyme]; thelysate was cleared by centrifugation following sonication (4 × 8 s)before incubation with 5 mL glutathione sepharose (GE Health-care) for 1 h. The beads were washed with 3 × 10 mL wash buffer(25 mM Tris pH 7.5, 0.5 M NaCl, 2 mM DTT, 0.5 mM EDTA),prior to immediate addition of lysate containing AKAP79.AKAP79 lysate was prepared by homogenization of frozen pelletfrom 6 L culture in 200mL (25 mMTris pH 7.5, 0.3MNaCl, 2 mMDTT, 0.5 mM EDTA, one EDTA-free protease inhibitor tablet)prior to centrifugation. After 1 h, the beads were washed with 3 ×10 mL low-salt wash buffer (as before with 300 mM NaCl). ForAKDPC reconstitution only, lysate from 12 L culture of PP2Bwas added. PP2B lysate was prepared as for D∕D using 200 mLmodified lysis buffer A (containing 0.3 not 0.5 M NaCl). After1 h incubation, the beads were washed with 3 × 10 mL low-saltwash buffer. In both cases, the beads were incubated with 1 mgPreScission protease in 6 mL low-salt wash buffer overnight.

In both cases, the eluate was collected in 16 mL low-salt washbuffer and desalted into nickel buffer A (25 mM Tris pH 8, 0.3 MNaCl, 15 mM imidazole) prior to 1 h incubation with 5 mL nickel-NTA beads (Qiagen). The beads were washed with 3 × 8 mLnickel buffer A, and AKAP79 and associated proteins wereeluted with 25 mM Tris pH 7.5, 0.3 M NaCl, 300 mM imidazole.Prior to concentration for AKDPC only, an excess of pure CaM(approximately 4 mg) was added to the eluate, and the buffer wassupplemented with 0.2 mM CaCl2. CaM was purified in advancefrom 6 L culture by glutathione affinity as for D∕D, followed byovernight PreScission cleavage and gel filtration in 25 mM TrispH 7.5, 0.5 M NaCl, 2 mM DTT, 0.5 mM EDTA. Finally, the el-uates were concentrated to 2 mL and applied to a Superdex S200gel filtration column with a 5 mLGSTtrap (both GEHealthcare).Gel filtration of AKAP79–D∕D was performed in 20 mM NaHepes pH 7.5, 200 mM NaCl—for AKDPC, the buffer was sup-plemented with 0.2 mM CaCl2. The peak fractions of AKDPC(64–68 mL) were pooled and applied once more to the gel filtra-tion column to reduce background contributed by aggregatedprotein in the void volume and excess PP2B (peak elution volumeapproximately 75 mL). All steps were performed at 4 °C.

Chemical Cross-Linking. N-terminal Strep(II)-tagged humanAKAP79 was expressed in Hi5 insect cells using the Bac-to-Bac Baculovirus System (Invitrogen). After baculoviral infection,

Hi5 cells were grown at 26 °C for approximately 72 h, then lysedby sonication in 50 mM Tris, pH 8.0, 500 mM NaCl, 2 mM DTT,1 mM EDTA, 1 mM benzamidine. Strep–AKAP79 was purifiedby affinity to streptactin resin (GE). Following elution with2.5 mM desthiobiotin, the protein was buffer exchanged into20 mM HEPES, 250 mM NaCl (pH 7.5) prior to cross-linking.A 100 mM stock of disuccinimidyl suberate (DSS) (Pierce Bio-technology Inc.) was prepared in anhydrous dimethylsulfoxide.The DSS stock was serially diluted and added to 0.5 mg∕mLpurified Strep–AKAP79 with a highest final concentration of1 mM DSS and allowed to react at room temperature for 30 min.Disulfide bonds were reduced through addition of tris(2-carbox-yethyl)phosphine (Pierce Biotechnology Inc.) to achieve a 5 mMfinal concentration and incubated at 56 °C for 1 h. Cysteineresidues were blocked by adding iodoacetamide (Sigma) to a finalconcentration of 10 mM and incubated at 37 °C for 30 min. Se-quencing grade trypsin (Promega) was added at 1∶250 ðwt∕wtÞand allowed to react overnight. The digestion reaction wasquenched by acidification using 0.1% TFA (Sigma-Aldrich).The sample was desalted using C18 Sepak (Waters Corporation).The sample was dried using a speed-vacuum device (ThermoScientific). Strong-cation exchange (SCX) separation of the pep-tides was performed using SCX spin columns (Nest Group).Mobile phase A was composed of 25% acetonitrile, 0.1% formicacid. Mobile phase B consisted of 1 M ammonium acetate(Sigma), 25% acetonitrile, and 0.1% formic acid. Peptides wereloaded in 100%mobile phase A and washed with mobile phase A.Eluted fractions include 3, 5, and 100% mobile phase B.

Liquid Chromatography and Mass Spectrometry of Cross-LinkedAKAP79. Liquid chromatography was performed using a WatersNanoAcquity UPLC (Waters Corporation). Pulled tip columnswere constructed in-house using a laser-pulling device (Sutter).A column of 30 cm in length was constructed with 75 μmID×360 μmOD fused silica capillary. The packing material usedfor peptide separation was 100-Å C18 magic beads (MicromBioresources). A fused silica trap column was constructed from100 μmID × 360 μmOD fused silica capillary. The frit was madeon one end of the trap with Kasil (PQ Corporation) to containC18 packing material. The packing material used in the trap was200-Å C18 magic beads (Microm Bioresources). A binary solventgradient was used for peptide separation. Mobile phase Aconsisted of 99.9% water with 0.1% formic acid. Mobile phaseB consisted of 95% acetonitrile with 0.1% formic acid and 4.9%water. The gradient was set up as follows: 5–60% B in 120 min.Column washing was done with 80% B for 20 min, followed byreequilibration for 20 min using 5% B.

MS was performed using a dual linear ion trap Fourier trans-form ion cyclotron resonance mass spectrometer, the Velos-FT,which is similar in performance to the Velos-Orbitrap (ThermoScientific). A precursor scan was acquired at 50,000 resolvingpower followed by 20 data-dependent tandem MS (MSMS)scans. Dynamic exclusion parameters were two repeat countsfollowed by 15-s repeat duration. An exclusion list size of 500species was used with exclusion of 30 s. Charge state screeningwas applied to exclude unassigned species, 1þ, and 2þ ions. Rawdata was converted using trans-proteomic pipeline tools (2) toMascot generic format.

Database Searching, Cross-Linked Peptide Identification, and AccurateMass Conformation. Data was searched using xQuest. A mass tol-erance of 10 ppm was used for precursor m∕z. The database

Gold et al. www.pnas.org/cgi/doi/10.1073/pnas.1014400108 1 of 9

http://www.pnas.org/cgi/doi/10.1073/pnas.1014400108

searched included only the AKAP79 amino acid sequence inFASTA format. For MSMS data, a tolerance of 0.1 Da was used.All hits were filtered using accurate mass and peptide MSMSdata. Homodimeric peptides are the only unambiguous cross-linked peptides useful for demonstration of protein–proteinproximity in the absence of other structural information aboutthe AKAP79 such as a crystal structure. Therefore, data wasfiltered for homodimeric peptide cross-links yielding three can-didates. Manual inspection of accurate mass, charge state, andfragmentation patterns allowed one site to be eliminated. In sam-ples of low complexity, accurate mass has been shown to allowunambiguous elemental composition, determination, and simpli-fied identification (3, 4). Here, we show that when AKAP79was digested in silico with up to two allowed missed cleavages,and all peptide cross-linked combinations were considered within10 ppm. The only cross-linked peptide candidates possible in thesample based on accurate mass measurements (Table S2) coin-cided with identifications obtained using the xQuest algorithm(Fig. S4). The xQuest search results reported are within accep-table limits for positive cross-linked peptide identification (5).

Protein Coimmunoprecipitation in HEK293 Cells. Two pcDNA3.1 vec-tors were constructed enabling transient expression of AKAP79,with either a FLAG or V5 tag at its N terminus, in HEK293 cells.The cells were harvested 2 d after transfection in 50 mM Tris pH7.5, 150 mM NaCl, 1% NP-40, 1 mM EDTA, 1 mM benzamidine,2 μg∕mL LeuPeptin, and FLAG–AKAP79 was imunoprecipitatedusing 4 μgmouse anti-FLAG antibody (Sigma). Immunprecipitationwith 4 μg mouse IgG was also attempted as a control. FLAG–

AKAP79 was detected using αFLAG–HRP conjugate (Sigma);V5–AKAP79 was detected using αV5–HRP conjugate (Invitrogen).

Pull-Down Assays. pGEX vectors were constructed for bacterialexpression of GST fused to either AKAP79WT, AKAP79Δ337–343, AKAP79 1–153 (“N”), AKAP79 154–296 (“M”) or AKAP79297–427 (“C”). The fusion proteins were expressed in E. coliBL21 cells by induction with 0.5 mM IPTG after the bacteriahad reached OD600 nm of 0.5. After 3 h, 1.8 L bacterial culturewas pelleted for each GST fusion protein and lysed in 100 mLlysis buffer A. The lysates were sufficient to saturate 300 μLglutathione sepharose beads. A FLAG tag was inserted at theC terminus of the A subunit of PP2B by site-directed mutagenesisin the bicistronic pGEX6P1 vector described earlier, and thetagged protein was purified as before.

In each pull-down experiment, 40 μL of glutathione sepharosebeads saturated with GST–AKAP79 fusion protein were used topull-down 2 μg PP2B–FLAG in 200 μL radioimmunoprecipita-tion assay (RIPA) buffer. When testing the effect of Ca2þ∕CaM,3 mM CaCl2 and 10 μg purified CaM was included. Following 2 hof incubation, the beads were washed in 5 × 1 mL RIPA buffer.RIPA buffer was supplemented with 3 mM CaCl2 when testingthe effect of Ca2þ∕CaM on pull-down. Beads were resuspendedin SDS-PAGE loading buffer and pull-down of PP2B–FLAG wasassessed by anti-FLAGwestern blot. For EDTA-mediated releaseexperiments, 50 μL of RIPA buffer supplemented with 3 mMEDTA and 3 mM EGTA was added, and after 30 min of incuba-tion, the supernatant fraction was separated from the sepharoseresin by centrifugation in a spin column.

Pull-Down Protein Phosphatase Assays. GST–PP2B was expressedfrom pGEX6P1 as for the AKDPC complex. Following enrich-ment by affinity to glutathione sepharose, PP2B was releasedby incubation with PreScission protease, further purified by affi-nity to calmodulin sepharose, and finally purified using a Super-dex S200 gel filtration column (pure untagged PP2B is shown inFig. S5B). Pull-down of 2 μg purified PP2B was attempted with40 μL of glutathione sepharose beads saturated with either GST,GST–AKAP79FL, or GST–AKAP79FLΔ337–343 in 250 μLRIPA buffer supplemented with 3 mM CaCl2 and 25 μg purifiedCaM. Following 2 h of incubation, pull-downs were washed in4 × 1 mL RIPA supplemented with 3 mM CaCl2. In one case,CaM was incubated with beads bearing GST–AKAP79FL for 2 h,and CaM not associated with AKAP79 was removed by washingwith 4 × 1 mL RIPA (þ3 mM CaCl2) prior to the PP2B incuba-tion step (Fig. 4D, lane 4). The beads were subsequently washedfor a second time in the normal way with 4 × 1 mLRIPA (þ3 mMCaCl2) to remove unbound PP2B. In another case, a single washstep with 1 mL RIPA buffer supplemented with 3 mM EDTA and3 mM EGTA was included following the final 4 × 1 mL washes inRIPA þ3 mM CaCl2 (Fig. 4D, lane 4). In every case, the beadswere subject to a final wash in 50 mM imidazole pH 7.2, 50 mMMgCl2, 5 mM NiCl2, 1 mM CaCl2 prior to phosphatase assay.

Phosphatase activity was measured using the Serine/ThreoninePhosphatase Assay System (Promega). Phosphate release wasquantified following 30 min incubation of pull-downs with100 μM RRA(pT)VA phosphopeptide in 50 mM imidazole pH7.2, 50 mM MgCl2, 5 mM NiCl2, 1 mM CaCl2.

Electrospray MS. Prior to MS analysis, AKAP79 complexes weredialyzed into 50 mM ammonium acetate pH 7.5� 0.1 M CaCl2.Mass spectra were obtained on a Q-Tof 2 or a Q-Tof 2B (Waters/Micromass), modified for high-mass operation (6), using a pre-viously described protocol to preserve noncovalent interactions(7). In brief, volatile aqueous buffered solutions were used withnanoelectrospray to ionize the complex and effect its transfer intothe gas phase intact. Additionally, careful optimization of instru-ment conditions, particularly the pressure and acceleratingvoltages, throughout the mass spectrometer was applied. Specifi-cally, to avoid defocussing of ions during their traversal through thevacuum, higher than standard pressures were applied to bringabout collisional focusing. The following experimental parameterswere used: capillary voltage 1.5 kV, sample cone 80–200 V,extractor cone 5 V, cone gas 100 L∕h, ion transfer stage pressure5 × 10−3–1.5 × 10−2 mbar, quadrupole analyzer pressure 1.0×10−5 mbar, and TOF analyzer pressure 8.1 × 10−6 mbar. The ac-celerating voltage into the collision cell was as low as 5 V fornonactivated MS spectra and was increased to 150 V for MSMSexperiments, with the collision cell pressurized to 35 μbar (argon).Spectra were externally calibrated using a 33 mg∕mL aqueous so-lution of cesium iodide (Sigma). Data were acquired and processedwith MassLynx software (Waters) and are shown with minimalsmoothing. For denaturing spectra, 50% acenonitrile (vol∕vol)and 0.1% formic acid (vol∕vol) were added to the purified samples

Structural Representation. Cartoon representations were gener-ated using PyMol (DeLano Scientific), and the electrostatic sur-face of PP2B was rendered using CCP4MG (8).

1. Mondragon A, et al. (1997) Overexpression and purification of human calcineurinalpha from Escherichia coli and assessment of catalytic functions of residues surround-ing the binuclear metal center. Biochemistry 36:4934–4942.

2. Pedrioli PG (2010) Trans-proteomic pipeline: A pipeline for proteomic analysis.Methods Mol Biol 604:213–238.

3. Smith RD, et al. (2002) An accurate mass tag strategy for quantitative and high-throughput proteome measurements. Proteomics 2:513–523.

4. Spengler B (2004) De novo sequencing, peptide composition analysis, and composi-tion-based sequencing: a new strategy employing accurate mass determination byfourier transform ion cyclotron resonancemass spectrometry. J Am SocMass Spectrom15:703–714.

5. Rinner O, et al. (2008) Identification of cross-linked peptides from large sequencedatabases. Nat Methods 5:315–318.

6. Sobott F, Hernandez H, McCammon MG, Tito MA, Robinson CV (2002) A tandem massspectrometer for improved transmission and analysis of large macromolecular assem-blies. Anal Chem 74:1402–1407.

7. Heck AJ (2008) Native mass spectrometry: A bridge between interactomics and struc-tural biology. Nat Methods 5:927–933.

8. Potterton E, McNicholas S, Krissinel E, Cowtan K, Noble M (2002) The CCP4 molecular-graphics project. Acta Crystallogr D 58:1955–1957.

9. Cole C, Barber JD, Barton GJ (2008) The Jpred 3 secondary structure prediction server.Nucleic Acids Res 36:W197–W201.



A

AKAP79-binding molecule Proposed basis of AKAP79 interaction Binding site within AKAP79

PKA Regulatory subunit D/D domain Amphipathic helix (387-406)

PP2B Strand 14 of PP2B A subunit 337PIAIIIT motif and surrounding region (315-360)

PKC Kinase domain substrate-binding site PKC pseudosubstrate sequence (31-52)

CaM Similar to MARCKS IQ motif (31-52)

PIP2 Inositol 4,5-bisphosphates Membrane-binding basic regions (MBBRs) (52-153)

AKAP79N C(427)

31 52 153 315 360 387 406

337PIAIIIT Tandem basic regions

AKAP79 JNET AKAP79AKAP75

mAKAP150rAKAP150


mAKAP150rAKAP150


mAKAP150rAKAP150


mAKAP150rAKAP150


mAKAP150rAKAP150


mAKAP150rAKAP150


mAKAP150rAKAP150

Basic region A31-52

RII-binding amphipathic helix387-406PxIxIT-like PP2B anchoring motif337-343

Cav1.2-binding site (partial leucine zipper)

Basic region B76-101

Basic region C116-145

B

Fig. S1. AKAP79 topology and annotated sequence alignment. (A) AKAP79 topology with color-matched table of binding partners. (B) Multiple sequencealignment of the human, bovine, mouse, and rat homologs of AKAP79 is shown. Above the alignment is a secondary structure prediction for AKAP79 gen-erated by JPRED (9) [arrows indicate regions of β-sheet secondary structure; cylinders indicate regions of helical secondary structure]. Previously mapped inter-action sites are indicated below the alignment. The three constructs used in this study, which were divided on the basis of previously mapped interactionmodules and sequence conservation, are designated by gray (N, 1–153), blue (M, 154–296), and red (C, 297–427) boxes.



A

B

2*(AKAP79+2*D/D)

A28

0nm

(mA

U)

Elution volume (ml)

Void volume

0

2

4

6

8

10

12

14

16

18

40 45 50 55 60 65 70 75 80 85 90 95 100 105 110

AKDPC complex

A28

0nm

(mA

U)

Elution volume (ml)

Free CaM

Fig. S2. Gel filtration of AKAP79 complexes. Elution profile absorbances at 280 nm for AKAP79 complex purifications using a Superdex S200 gel filtrationcolumn are shown. (A) AKAP79–D∕D complex elution. The peak at 45.7 mL corresponds to protein aggregates eluting in the void volume; the peak at 66.6 mL isthe AKAP79–D∕D complex (B) AKDPC complex elution. The second peak at 96 mL corresponds to excess CaM.



Fig. S3. MS andMSMS spectra of AKAP79–D∕D complexes. (A) TheMS spectrum of the AKAP79–D∕D complex at very low activation energies, which minimizedissociation, is shown. Peak series for the [AKAP79þ 2 � D∕D] and [2 � AKAP79þ 2 �D∕D] complexes used for MSMS are indicated. (B) MS spectrum of theAKAP79–D∕D complex under denaturing conditions. This approach enabled a highly accurate mass to be calculated for AKAP79. (C) MSMS spectrum ofAKAP79–D∕D where ions at 3,640 m∕z were selected and activated by collision-induced dissociation. Activation resulted in the dissociation of a D∕D subunitwith charge states from 3þ to 6þ and the formation of the corresponding stripped complex of 56,302 Da [1 � AKAP79þ 1 � D∕D] (gold). Upon further activa-tion, this complex expelled a second D∕D subunit to form a stripped complex of 50,801 Da [monomeric AKAP79] (blue). Thus, the original precursor ion can beassigned to the 17þ charge state of a complex composed of [1 � AKAP79þ 2 � D∕D], one AKAP79 molecule binding to two D∕D subunits. (D) Strep–AKAP79dimer cross-linking with 0 (lane 1), 50 (lane 2), and 500 (lane 3) μM DSS, as shown by Coomassie staining.



Fig. S4. Identification of dimeric AKAP79 cross-linking sites with xQuest. The xQuest scoring data for peptides containing (A) K328–K328 cross-links and (B)K333–K333 cross-links are shown. Reported scores ≥8 are generally considered to be significant.



IB: PP2B

25

20

kD

Ca2+ / CaM:

Bait: GST | GST-AKAP79 | GST-AKAP79NFAT

75

50

Input

Coomassie:

0

0.5

1

1 2 3 4 5 6

Supplementary PP2B-FLAG pull-down experiments

C

D

0

0.5

1

1 2

Ca2+:

CaM:

IB: FLAG

IB: CaM

Coomassie50

37

kD

Bait: GST-AKAP79 (1-153)

50

37

25

20

75

15

kD

10

FLAG-PP2BA B Untagged PP2B

50

37

25

20

75

15

kD

10

Fig. S5. PP2B preparations and additional PP2B–FLAG pull-down experiments. Coomassie-stained 4–12% SDS-PAGE of (A) 4 μg purified PP2B–FLAG and(B) 2 μg untagged PP2B. (C) Anti-FLAG immunoblot shows PP2B–FLAG pulled-down with GST–AKAP79 (1–153) and either 3 mM Ca2þ alone (lane 1) or3 mM Ca2þ in the presence of 10 μg CaM (lane 2). Relative densitometry analysis (n ¼ 3, Top), anti-calmodulin IB (Middle) and pull-down Coomassie (Bottom)are shown. These data demonstrate that PP2B interaction with the AKAP79 N-terminal site requires CaM. (D) Pull-down of PP2B–FLAG was attempted�Ca2þ∕CaM with either GST, GST–AKAP79, or GST fused to a mutant of AKAP79 in which the PIAIIIT motif was replaced with the sequence PRIEIT “AKAP79N-FAT.” The inability of AKAP79NFAT to bind PP2B in the absence of Ca2þ∕CaM suggests that the PIAIIIT motif cooperates with other sequence determinants inthe C-terminus of AKAP79 to anchor PP2B. Specifically, switching the anchor residues Pro337 and Thr343 to opposite sides of the anchoring-sheet may beincompatible with other binding motifs in the C-terminus of AKAP79.



Asymmetrical Processive

AKAP79

PP2B

B BCaM

[Ca2+]

B

CaM

PP2B

[Ca2+]

CaM + PP2B

A

Asymmetrical Non-Processive

[Ca2+]

[Ca2+]

CaM +

PP2B

AKAP79

B

CaM +

PP2B

Symmetrical

A

B [Ca2+]

[Ca2+]CaM

CaM

PP2B

CaM

A

B

BAKAP79

PP2B

CaM

In each case the second AKAP79 protomer and associated proteins are not illustrated for clarity

B

C

AKAP79-PP2B Anchoring Models

Fig. S6. Models of AKAP79–PP2B anchoring. In each model, site A corresponds to the PIAIIIT motif, and site B corresponds to the N-terminal Ca2þ∕CaM-sensitive site. (A) Symmetrical. In this model two PP2B heterodimers associate in parallel on either side of the PIAIIIT motif of site A. Upon CaM activation,elements in the PP2B B subunits and/or P2B A subunit C-terminal regions of each anchored heterodimer simultaneously engage AKAP79. (B) Asymmetricalnonprocessive. In this model, the A site is constitutively occupied by one PP2B heterodimer, whereas another PP2B heterodimer can associate and dissociatewith site B under the influence of Ca2þ∕CaM. (C) In this model, similar to the asymmetrical nonprocessive model, AKAP79 contains two separate PP2B bindingsites. In the absence of Ca2þ∕CaM, one PP2B heterodimer associates with site A, where it is primed for translocation to site B upon CaM activation. UponCa2þ∕CaM-mediated translocation of PP2B from site A to site B, site A becomes unoccupied, allowing association with a second PP2B heterodimer. Once cellular[Ca2þ] decreases sufficiently, both CaM and the PP2B heterodimer associated with the B site will dissociate, returning the signaling complex to its ground statewith one PP2B heterodimer bound to the AKAP79 A site. It is important to note that for each of these models only one protomer of the AKAP79 dimer isdescribed for the sake of clarity. The assumption is that each protomer in the AKAP79 dimer anchors PP2B in the same way.

Table S1. Theoretical and measured masses of AKAP79 signaling complexes and substituent proteins

Component protein, s Theoretical mass, Da Experimental mass Da Deviation, %

RIIα D∕D 5,624 5,627 0.05PP2B A subunit 58,070 58,072 0.003PP2B B subunit (5–170) 18,915 18,915 0AKAP79 50,460 50,461 0.002CaM 17,249 17,321 0.42AKAP79þ D∕D 56,084 56,302 0.39AKAP79þ 2 � D∕D 61,708 61,929 0.362 � AKAP79 100,920 101,036 0.112 � AKAP79þ 1 � D∕D 106,544 106,820 0.262 � AKAP79þ 2 � D∕D 112,168 112,447 0.25AKDPC complex 2 � ½AKAP79þ 2 � PP2BðAþ BÞ þ 2 � D∕Dþ CaM� 465,854 466,051 0.04

Predicted masses and masses recorded by MS of AKAP79 complex-associated proteins are listed.



Table S2. DSS cross-linked peptide candidates

Peptide 1 Peptide 2 Neutral mass (peptide1 + peptide2 + 138.0680796 Measured mass PPM error

1,821.91507SKSEESK GAWASLKR 1,818.947578 1,628.78SKNVPK GPKRSNHSK 1,819.00643 1,596.47AAKALKPK FPRGPKR 1,820.114858 988.09SKNVPK ASMLCFKR 1,820.964096 521.96SEESKR SKSEESKR 1,821.906835 4.52KAAK EKASMLCFKR 1,822.979746 −584.37KAAK LKIPCIKFPR 1,825.101182 −1748.77SEESKR KAAKALKPK 1,826.062548 −2276.44

2,037.02797GAWASLKR GPKRSNHSK 2,035.107541 942.76SKSEESKR SKSEESKR 2,037.033826 −2.87GAWASLKR ASMLCFKR 2,037.065207 −18.28KAAK ALKPKAGSEAADVAR 2,037.158239 −63.95QKEK QKEKASMLCFK 2,038.059119 −506.20LKIPCIK IPCIKFPR 2,038.183532 −567.28KAKSR IPCIKFPRGPK 2,038.187371 −569.16

AKAP79 was digested with trypsin in silico with up to two allowed missed cleavages. Potential cross-linked AKAP79 peptidesequences within 10 ppm of two measured masses are listed. The best solutions are in bold.

Table S3. The best stoichiometric assignments for the 466-kDa AKDPC complex

Stoichiometric assignment* Estimated mass, Da

(R_II_(1–45))4_(CaM)2_(PP2B_B)4_(AKAP_79)2_(PP2B_A_)4 465,854(R_II_(1–45))4_(CaM)11_(PP2B_B)5_(AKAP_79)2_(PP2B_A_) 465,800(R_II_(1–45))4_(CaM)3_(AKAP_79)2_(PP2B_A_)5 465,513(R_II_(1–45))4_(CaM)12_(PP2B_B)_(AKAP_79)2_(PP2B_A_)2 465,459(R_II_(1–45))4_(PP2B_B)15_(AKAP_79)2_(PP2B_A_) 465,211(R_II_(1–45))4_CaM_(PP2B_B)11_(AKAP_79)2_(PP2B_A_)2 464,870(R_II_(1–45))4_(CaM)2_(PP2B_B)7_(AKAP_79)2_(PP2B_A_)3 464,529(R_II_(1–45))4_(CaM)11_(PP2B_B)8_(AKAP_79)2 464,475(R_II_(1–45))4_(CaM)3_(PP2B_B)3_(AKAP_79)2_(PP2B_A_)4 464,188(R_II_(1–45))4_(CaM)12_(PP2B_B)4_(AKAP_79)2_(PP2B_A_) 464,134(R_II_(1–45))4_(PP2B_B)18_(AKAP_79)2 463,886(R_II_(1–45))4_(CaM)13_(AKAP_79)2_(PP2B_A_)2 463,793

To determine the most likely stoichiometry of the 466,051-Da AKDPC complex, masses werecalculated for all potential combinations of interactions between AKAP79 signaling complexcomponent proteins, based on their theoretically predicted masses (see Table S1), usingsoftware developed by ref. 1. We assumed that an AKAP79 dimer associated with 4 � D∕D isthe basis of the complex (refer to Fig. 1). Solutions within 0.5% of the measured mass areshown; the closest solution to the measured mass is in bold. With the exception of the bestsolution, 2 � ½AKAP79þ 2PP2BðAþ BÞ þ 2RIIðD∕DÞ þ CaM�, the stoichiometric solutions do notconform to either the known 1∶1 ratio of the PP2B heterodimer, or to liberal ratios of[n � CaM ≤ 3n � AKAP79þ n � PP2B].*Assuming error margins of 0.5% and AKAP: RII D∕D stoichiometry of 2∶4 based on MSMS data.

1 Rinner O, et al. (2008) Identification of cross-linked peptides from large sequence databases. Nat Methods 5:315–318.



supporting information - university of...

Documents